We have to know where an object is in visual space relative to the body, in order to look at it, move towards it or away from it, reach out and touch it, grasp it, or even throw something at it. Visual information enters the brain via the retina, but because the retina moves in the orbit, knowing where something is on the retina does not automatically tell the brain where it is in space. In order to calculate where something is in space, the brain must know not only where it is in the retina, but where the retina is in space. The brain knows where the head is in space from signals from the vestibular system in the inner ear, and from sensors in the neck muscles. There are signals in the visual association cortex of the brain that report where the eye is in the orbit, but the source of this eye position information is unknown. One possibility is that the source of the eye position signal arises as a corollary discharge from the motor system, because the signal controlling the muscles that move the eye has a component signaling the position of the eye. We have previously discovered such a corollary signal which reports the dimensions of an impending eye movement to cortical neurons. The other possibility is that sensors in the extraocular muscles send a signal to the brain describing the position of the eye in the orbit. Monkey and human extraocular muscles have many sensors which look like the sensors in the skeletal muscles that signal muscle length. These sensors project through the thalamus to area 3a in primary somatosensory cortex (SI), and to the second somatosensory cortex (Sll). In very preliminary experiments we have found cells in both area 3a and Sll (as identified by magnetic resonance imaging) which describe the position of the eye in the orbit. This proposal has two aims: 1) to characterize the eye position signal, looking at its oculomotor, visual, and attentional properties. 2) to determine by anesthetizing the orbit of one eye, if the signal has a proprioceptive or a corollary origin. These experiments will provide insight into the mechanisms by which the brain analyzes space for perception and action, and will help us understand and design rehabilitative strategies for deficits in spatial behavior such as occur after strokes involving the parietal cortex, of which SI and Sll form a component. [unreadable] [unreadable] [unreadable]